Conceptualizing the relationship between chronic migraine and episodic migraine

Migraine is commonly divided into two broad forms, episodic migraine (EM) and chronic migraine (CM). EM occurs less than 15 days per month and is characterized by attacks of pain and associated symptoms, including nausea, photophobia, phonophobia and aura in various combinations. According to the International Classification of Headache Disorders, Second Edition (ICHD-2), CM, is characterized by headaches that occur on 15 or more days per month; on at least 8 of those days the headaches must either meet diagnostic criteria for migraine without aura or respond to migraine-specific treatment [1]. In ICHD-2, EM and CM are distinguished by a clearly defined, but somewhat arbitrary, boundary of headache frequency.

Efforts to conceptualize the relationship between CM and EM are rare. ICHD-2 conceptualizes CM as a complication of EM, viewing it as a form of very high frequency EM. An alternative view suggests that CM and EM are qualitatively distinct, that they differ not just in degree but also in class. This perspective postulates that CM may involve brain mechanisms distinct from those of EM. Herein, we argue that CM and EM should be conceptualized as distinct but intimately related disorders.

Delineating the boundary between related conditions is a classic problem in classification research [2]. Ideally, disease classification is based on causes and mechanisms (etiologic classification), but when etiologic classification is not possible symptom profiles are used (symptomatic classification). In the ICHD-2, secondary headache disorders are classified based on their etiology as headaches attributable to an underlying cause, such as an infection or a mass lesion. For primary headache disorders, such as CM and EM, the exact etiology is unknown and so diagnostic boundaries must be defined in other ways.

In the absence of an etiologic gold standard, conditions can be distinguished based on a series of features, including symptoms and signs, clinical course/prognosis, epidemiologic profiles, comorbidities, treatment response and family history/genetics, as well as neuroimaging findings and biomarkers [2]. No single feature can be used to reliably determine whether EM and CM differ simply in degree or also in kind. These judgments rely on the overall pattern of features and ultimately on insights into the etiopathogenesis as they emerge. We will discuss CM and EM taking these features into account (Table 1).

Chronic migraine and EM differ in attack frequency and in clinical course. In general, EM is characterized by a relatively stable course with a headache frequency from 1 to 4 days per month ranging up to 14 days per month [3,4]. By contrast, CM is characterized by a phase of headache acceleration followed by a plateau phase with attacks occurring from 15 to 30 days per month [4–6]. CM is far more disabling than EM and more likely to be persistent [6–8].

Chronic migraine and EM also differ in their epidemiologic profiles, risk factors and comorbidities. In the general population, individuals with CM are older, have lower levels of education, lower household income and are less likely to be working full time [9–11]. Risk factors for EM and CM overlap but differ in important ways. For example, female gender is a risk factor for the new onset of EM but it is not a risk factor for progression to CM [9,12,13]. Obesity is not a risk factor for migraine onset but is a risk factor for CM onset among suffers of EM [14–16]. Most of the well-known comorbidities of migraine are significantly more common in people with CM. These include several psychiatric comorbidities (depression and anxiety), all forms of chronic pain and many cardiovascular disorders [17–20].

Cutaneous allodynia (CA) is defined as the experience of pain provoked by stimulation of the skin that would ordinarily not produce pain. Clinic- and population-based studies suggest that approximately two thirds of patients with migraine have CA and that CA is more common in CM than in EM [21–24]. The mechanism of facial CA is central sensitization of the nociceptive neurons in the trigeminal nucleus caudalis, which receives convergent afferent input from the dura mater and periorbital skin [25,26]. Repetitive activation of trigeminovascular neurons and consequent repetitive activation of modulatory pain pathways involving the periaqueductal gray may lead to structural changes or functional impairment of neurons in the periaqueductal gray [27–29]. In other words, the experience of repetitive headaches may lead to functional and structural brain changes that set the stage for more headaches.

While there is considerable overlap in treatment response for CM and EM, differences are emerging. Triptans treat pain exacerbations for both EM and CM [30,31]. Some preventive medications, including topiramate, have demonstrated effectiveness in both groups of patients [32–34]. Emerging evidence suggests that botulinum toxin is not effective in EM but is effective in CM [35,36].

The role of family history in CM versus EM is uncertain. EM aggregates within families and is being considered a complex genetic disorder; multiple genetic and environmental factors contribute to the development of EM [37–39]. For CM, the role of family history is not well known. One epidemiologic study reported that chronic daily headache aggregates within families of probands with chronic daily headache to a greater degree than in the families of probands with episodic headache [17].

There are quantitative and qualitative differences in brain structure and function between EM and CM, although the study of this area is in its infancy. Cross-sectional studies show that the number of brain lesions in the deep white matter and the number of small stroke-like lesions in the posterior circulation increase with the frequency of migraine attacks, particularly for patients with migraine with aura [40,41]. Iron deposition in the substania nigra and red nucleus, which are involved in central pain processing, was significantly more prominent in an MRI study of CM than in EM [29]. These patterns suggest that brain lesions may be a consequence of multiple attacks. As brain lesions are most strongly associated with migraine with aura and increase in frequency with the number of attacks, the mechanism of aura may be linked to the mechanisms of anatomic change. Many lines of evidence suggest that cortical spreading depression, the substrate of the migraine aura, may contribute directly to brain injury in migraine with aura [42–44].

Physiological or functional differences between CM and EM also have been reported. CM showed significantly lower phosphene thresholds in transcranial magnetic stimulation (TMS) compared with EM [45]. Another test for cortical excitability is magnetic suppression of perceptual accuracy, which examines the effects of TMS on visual perception. In CM, the accuracy of letter reporting pattern was not influenced by TMS, but this was a distinct pattern from EM in which the accuracy of letter reporting was decreased at the mid-range interval by TMS [45,46]. These data suggest that the cortical excitability in CM is more prominent than in EM.

Migraine is a chronic disorder with episodic attacks. In some individuals migraine may progress, giving rise to CM. Although CM is classified as a complication of EM in the ICHD-2, herein we argue that it is a distinct disorder with a clinical profile, prognosis, epidemiological profile and pattern of treatment response that differs from EM. Recognizing these differences and clarifying mechanistic differences may lead to distinct treatments for CM and EM. Given the debilitating nature of CM, developing strategies to prevent and treat this disorder are high priority public health goals.

Table 1. Contrasting features of episodic and chronic migraine.

Feature	Episodic migraine	Chronic migraine
Symptoms and signs	Less disability, less allodynia	More disability, more allodynia
Clinical course/prognosis	Attacks less frequent and more like to remit	Attacks more frequent and less likely to remit
Epidemiological profile	Younger, inverse relation to socioeconomic status	Older, inverse relation to socioeconomic status
Comorbidities	Many comorbidities, less strongly associated	Psychiatric and medical comorbidities, more strongly associated with CM than EM
Treatment response: acute	Triptans, NSAIDs	Treatment response less robust to triptans;Early treatment is more important;Risk of medication overuse is greater
Treatment response: preventive	AEDs and b blockers are effective; botulinum toxin is ineffective	Topiramate is effective; botulinum toxin is effective.
Family history/genetics	Family relative risk for EM approximately double	CDH may aggregate in families
Structural and functional findings	DWMLs and posterior circulation strokes in migraine with aura	More DWMLs, iron deposition increases with frequency, related to stroke^*; cortical excitability is more prominent in CM than EM
Biomarkers	No biomarker is currently available for EM	No biomarker is currently available for CM

^*Unpublished data from the American Migraine Prevalence and Prevention Study.

AED: Antiepileptic drug; CDH: Chronic daily headache; CM: Chronic migraine; DWML: Diffuse white matter lesion; EM: Episodic migraine.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.